Technological Competencies in the
World's Largest Firms:
Characteristics, Constraints and
Scope for Managerial Choice
Patel, P. and Pavitt, K.
IIASA Working Paper
WP-95-066
July 1995
Patel, P. and Pavitt, K. (1995) Technological Competencies in the World's Largest Firms: Characteristics, Constraints and
Scope for Managerial Choice. IIASA Working Paper. IIASA, Laxenburg, Austria, WP-95-066 Copyright © 1995 by the
author(s). http://pure.iiasa.ac.at/4528/
Working Papers on work of the International Institute for Applied Systems Analysis receive only limited review. Views or
opinions expressed herein do not necessarily represent those of the Institute, its National Member Organizations, or other
organizations supporting the work. All rights reserved. Permission to make digital or hard copies of all or part of this work
for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial
advantage. All copies must bear this notice and the full citation on the first page. For other purposes, to republish, to post on
servers or to redistribute to lists, permission must be sought by contacting [email protected]
Working Paper
Technological Competencies in the
World's Largest Firms:
Characteristics, Constraints and
Scope for Managerial Choice
Pari Patel and Keith Pavitt
WP-95-66
July 1995
0
' 1 lASA
IL
AI
D~HHD
International Institute for Applied Systems Analysis
Telephone: +43 2236 807
A-2361 Laxenburg
Telefax: +43 2236 71313
Austria
E-Mail: [email protected]
Technological Competencies in the
World's Largest Firms:
Characteristics, Constraints and
Scope for Managerial Choice
Pari Patel and Keith Pavitt
Science Policy Research Unit
University of Sussex
WP-95-66
July 1995
This paper is based on research at the Centre for Science, Technology and
Energy and Environment Policy (STEEP), funded by the Economic and
Social Research Council (ESRC) within the Science Policy Research Unit.
We have benefited greatly from comments and criticisms following
presentations at the Academic Commission on Externalities and the
Performance of Firms (Paris, January, 1994), and at the Fourth International
Conference on the Management of Technology (Miami, March, 1994).
Working Papers are interim reports on work of the International Institute for Applied
Systems Analysis and have received only limited review. Views or opinions expressed
herein do not necessarily represent those of the Institute or of its National Member
Organizations.
DL
AD
B M D ~
lASA
International Institute for Applied Systems Analysis
Telephone: +43 2236 807
A-2361 Laxenburg
Telefax: +43 2236 71313
.
Austria
E-Mail: [email protected]
Preface
The research project on Systems Analysis of Technological and Economic Dynamics a t IIASA is
concerned with modeling technological and organisational change; the broader economic developments that are associated with technological change, both as cause and effect; the processes
by which economic agents - first of all, business firms - acquire and develop the capabilities
to generate, imitate and adopt technological and organisational innovations; and the aggregate
dynamics - at the levels of single industries and whole economies - engendered by the interactions among agents which are heterogeneous in their innovative abilities, behavioural rules and
expectations. The central purpose is to develop stronger theory and better modeling techniques.
However, the basic philosophy is that such theoretical and modeling work is most fruitful when
attention is paid to the known empirical details of the phenomena the work aims to address:
therefore, a considerable effort is put into a better understanding of the 'stylized facts' concerning corporate organisation routines and strategy; industrial evolution and the 'demography' of
firms; patterns of macroeconomic growth and trade.
From a modeling perspective, over the last decade considerable progress has been made on
various techniques of dynamic modeling. Some of this work has employed ordinary differential
and difference equations, and some of it stochastic equations. A number of efforts have taken
advantage of the growing power of simulation techniques. Others have employed more traditional
mathematics. As a result of this theoretical work, the toolkit for modeling technological and
economic dynamics is significantly richer than it was a decade ago.
During the same period, there have been major advances in the empirical understanding.
There are now many more detailed technological histories available. Much more is known about
the similarities and differences of technical advance in different fields and industries and there is
some understanding of the key variables that lie behind those differences. A number of studies
have provided rich information about how industry structure co-evolves with technology. In
addition to empirical work a t the technology or sector level, the last decade has also seen a
great deal of empirical research on productivity growth and measured technical advance a t the
level of whole economies. A considerable body of empirical research now exists on the facts that
seem associated with different rates of productivity growth across the range of nations, with the
dynamics of convergence and divergence in the levels and rates of growth of income, with the
diverse national institutional arrangements in which technological change is embedded.
As a result of this recent empirical work, the questions that successful theory and useful
modeling techniques ought t o address now are much more clearly defined. The theoretical work
has often been undertaken in appreciation of certain stylized facts that needed t o be explained.
The list of these 'facts' is indeed very long, ranging from the microeconomic evidence concerning
for example dynamic increasing returns in learning activities or the persistence of particular sets
of problem-solving routines within business firms; the industry-level evidence on entry, exit and
size-distributions - approximately log-normal- all the way to the evidence regarding the timeseries properties of major economic aggregates. However, the connection between the theoretical
work and the empirical phenomena has so far not been very close. The philosophy of this project
is that the chances of developing powerful new theory and useful new analytical techniques can
be greatly enhanced by performing the work in an environment where scholars who understand
the empirical phenomena provide questions and challenges for the theorists and their work.
In particular, the project is meant t o pursue an 'evolutionary' interpretation of technological
and economic dynamics modeling, first, the processes by which individual agents and organisations learn, search, adapt; second, the economic analogues of 'natural selection' by which inter-
active environments - often markets - winnow out a population whose members have different
attributes and behavioural traits; and, third, the collective emergence of statistical patterns,
regularities and higher-level structures as the aggregate outcomes of the two former processes.
Together with a group of researchers located permanently a t IIASA, the project coordinates
multiple research efforts undertaken in several institutions around the world, organises workshops
and provides a venue of scientific discussion among scholars working on evolutionary modeling,
computer simulation and non-linear dynamical systems.
The research focuses upon the following three major areas:
1. Learning Processes and Organisational Competence.
2. Technological and Industrial Dynamics
3. Innovation, Competition and Macrodynamics
SUMMARY
Firm-specific technological competencies are major factors explaining why firms are
different, how they change over time, and whether or not they are capable of remaining
competitive. Systematic analysis of the technological activities of more than 400 of the
world's largest firms shows that their technological competencies have the following
characteristics.
They are highly diversified. Large firms are typically multi-technology. The most
pervasive competencies remain in mechanical, chemical and instrumentation
engineering, and with an increasing spread of competencies in computers, materials
and biotechnology.
They are highly stable and differentiated in composition, with both the technology
mix and the directions of localised search strongly influenced by the firm's principal
products.
The rate of search (as measured by the level and rate of increase of total innovative
activities, and by the rate of entry into fast-growing technical sub-fields) is influenced
by both the firm's principal products, and the conditions in its home country.
However, considerable unexplained variance suggests scope for managerial choice
in the overall commitment of resources to the accumulation of technological
competencies, and in the vigour with which promising sub-fields are explored.
These findings:
1. confirm the importance of complexity and path dependency in the accumulation of
firm-specific technological competencies;
2. demonstrate that technological competencies give a convincing empirical explanation
of the boundaries (or - and perhaps better - the core activities) of firms.
3 . challenge many of the standard taxonomies of technology strategies in large firms. In
particular:
- firms' technological diversity challenges notions of "focus", "core competence",
"competence-destroying innovations", and "technological leap-frogging";
- firms' differentiated competencies and path dependency put severe limits on the
range of exploitable technological opportunities;
- firms' stability in technology mix shows that technological accumulation and
change are slow processes.
4. confirm the importance in technology strategy of integration (or "fusion") of different
fields of technological competence.
5. point towards the importance of complementary managerial competencies in
organisational integration, methods of resource allocation, and learning.
Technological Competencies in the
World's Largest Firms:
Characteristics, Constraints and
Scope for Managerial Choice
Pari Patel and Keith Pavitt
1 INTRODUCTION
1.1 Why Firm-Specific Technological Competencies are Important
The purpose of this paper is to throw empirical light on the nature and determinants of the
technological competencies of the world's largest firms.
The subject of "firm-specific
competencies" is of increasing interest to practitioners, and to theorists - and particularly to
those in the neo-Schumpeterian tradition, who are seeking to explain why firms provide
different ranges of goods and services, why they change at different rates and in different
directions over time, and what makes them competitive (Rumelt, 1974; Ramanujam and
Varadarajan, 1989; Prahalad and Hamel, 1990; Dosi et al., 1992; Carlsson and Eliasson, 1991;
Teece et al., 1992, Teece et al., 1993).
Our main data source is systematic information of US patenting by more than 400 of the
world's largest technologically active firms, broken down by each firm's nationality
(headquarters country) and principal product group, and by the technical field and by the
country of origin of the inventor of each patent'. Similar data has been used by Hall and her
colleagues (1986) to measure lags between R & D and patenting at the firm level, by Narin
and his colleagues (1987, 1988) for corporate and competitor analysis, by Jaffe (1986, 1989)
to identify and measure technological "spillovers", and by Cantwell (1991) to explain patterns
of international production.
I These firms have been chosen from the list of world's largest firms published in the Fortune magazine in 1988.
Only firms with more than 50 patents granted in the US in the period 1981-90 have been included. For a detailed
description of the characteristics and method of compilation of the database see Patel and Pavitt (1991).
1.2 The Main Questions and their Answers
We concentrate here on systematic comparisons of the level, rates of change and composition
(by technical field) of each firm's patenting activity, and on their characteristics and
determinants. In this paper, our level of analysis is not detailed enough either to identify a
specific company's distinctive competence within a product field, or to describe how it
accumulates technology to gain competitive advantage2. Instead, we intend to answer two
questions.
First: "What are the characteristics of technological competencies in large firms?" We shall
show that they are:
diversified (i.e. multi-technology) and evolving over time;
heavily differentiated and stable in their composition and their directions of search,
both as a function of the products that they make.
Second: "What are the constraints on the development of technological competencies in large
firms, and what in consequence is the scope for managerial choice?" We shall show that:
the rate of search is significantly influenced by both the firm's product mix and country
of origin;
there is considerable unexplained variance in the aggregate level of technological
activity, and in the rate of entry into fast-growing technical sub-fields.
This suggests that, whilst directions of search are heavily constrained by accumulated
competencies, considerable scope for managerial choice remains in fixing the rate of search.
1.3 The Framework of Explanation: Coping with Complexity
Both our questions and our answers are consistent with the neo-Schumpeterian framework of
analysis, based on the pioneering work of Nelson and Winter (1982)3. Technological
artefacts, and the organisational and economic worlds in which they are embedded, are
complex: in other words, they each comprise so many variables and interactions that it is
impossible fully to model, predict and control their behaviour through explicit and codified
theories and guidelines. Certainty about the future, probabilistic risk and optimisation are
therefore impossible. The best approach to problem-solving and the management of change is
step-by-step experimentation, in which changes are made in one feature or component at a
For a recent example of the latter, see Miyazaki (1994), who used bibliometrics and interviews to trace how a
number of major companies assimilated opto-electronics technologies. She found cumulative paths of learning:
directions of search were influenced by previously accumulated competencies; and over time search became
more focused and applied.
See also Rumelt et al., 199 1.
time, and ends and means re-interpreted in the light of the subsequently observed changes. In
addition to codified knowledge, experience and tacit knowledge improve the effectiveness of:
the choices of the feature or component to vary at each stage;
subsequent modifications in means and ends made after observation of the effects of
variations in features or components.
This method is called "learning", or "experimentation", or "trial and error" (and many other
things, including "suck it and see").
Essentially the same approach underlies Lindblom's
prescriptions in public policy (1959), Quinn's in corporate strategy (1980), and Kline's in
engineering design and development (1990). It explains our results, as follows:
the complex and multivariate nature of technological artefacts requires the combination
and application of advances in many fields of knowledge: hence large firms'
competencies are typically multi-technology, and evolving over time;
complexity also constrains firms to search and experiment in and around what they
already know and produce: hence firms competencies are differentiated, stable, and
closely related to their product mix;
the rate and direction of a firm's search will be influenced by the opportunities and
incentives that it faces. These will depend on its own accumulated competencies, and
on its surrounding environment: hence the influence of both principal product group,
and home country on firms' level of technological activities;
but complexity means uncertainty, and the impossibility for a firm to identify all
possible future states, let alone to predict what will happen. It also means difficulty
and uncertainty in identifying the competitive competencies that the firm has at its
disposal. Hence the unexplained variance in the level of technological activities and
in the rate of entry in fast-growing sub-fields, reflecting the scope for managerial
choice.
1.4 Limitations of our Analysis
Our paper has three sets of limitations. First, we measure only technological competence, and
thereby neglect many others that are important. Dosi and Teece(1993) have distinguished
organisational-economic competencies from technical competencies, and have argued that the
latter derives from the former, and is therefore more fundamental to the firm4. Our empirical
"Organisational/economic competence involves: (1) allocative competence - deciding what to produce and
how to price it; (2) transactional competence - deciding whether to make or buy, and whether to do so alone or
in partnership; and (3) administrative competence - how to design organisational structures and policies to
enable efficient performance. Technical competence, on the other hand, includes the ability to develop and
design products and processes, and to operate facilities effectively .....................A firm becomes superior in a
results suggest that this is only partly correct. A firm's organisational competence does
influence its level of commitment to technological activities, and its rate of entry into fastgrowing sub-fields. However, a firm's accumulated technological competence strongly
constrains the directions in which it searches: even the brightest and the best organisational
capabilities will find it difficult (impossible?) to convert a firm making Harris Tweed jackets,
or Italian high-fashion shoes, into a world class firm in personal computers.
The
differentiated nature of technical competencies is one the most important factors explaining
the coherence and the boundaries of the firm. And a recent survey of 100 Italian firms by
Malerba and Marengo (1993), ranked technological competencies as of greater long term
importance than competencies to respond to either market signals or competitors' strategic
actions. The subject therefore deserves analytical and empirical attention, even if it does not
cover - and cannot explain - everythings.
The second limitation is that we measure technological competencies only imperfectly
through patent data6. Nonetheless, patenting in the USA is a better measure than most, if not
all, the alternatives given its relative homogeneity, detail, accuracy and (after recent advances
in information technology) accessibility and cost: hence its increasing use by both analysts
and practitioners7. However, in relation to the subject of this paper, three potential limitations
of the US patenting measure must be mentioned:
1. Patents do not measure the extent of the firm's external technological linkages.
However, many studies have shown (most notably, Cohen and Levinthal, 1989) that
external technological linkages are in general complementary to internal competencies,
and these we do measure.
2. Patents measure codified knowledge, whereas a high proportion of firm-specific
competencies is non-codified (i.e. tacit) knowledge. We would argue that the two
forms of knowledge are complementary, not substitutes. Other measures that embody
tacit knowledge (such as R & D expenditure, judgements of technological peers) give
results very similar to those using patenting (see Pate1 and Pavitt, 1987).
particular technological domain because it has certain organisational capabilities: it allocates resources to more
promising projects, it harnesses experience from prior projects, it hires and upgrades human resources, it
integrates new findings from external sources, and it manages a set of problem-solving activities associated with
that technology." (Dosi and Teece, 1993, pp. 6-7).
In a similar manner (and using the jargon of another academic discipline), we are fully aware that
technological competencies in large firms are "socially constructed (Hughes and Pinch, 1987). But we
concentrate here on the important cognitive factors that shape the social construction of technology.
The uses and abuses of patent data have been extensively discussed elsewhere, See, for example, Basberg,
1987; Pavitt, 1988; Grilliches, 1990; Patel and Pavitt, 1994a.
In addition to Jaffe (1986, 1989) and Narin And Noma (1988) see - for example - Griliches (1984) and
Business Week, (1993).
3. Patenting does not fully measure competencies in software technology, since copyright
law is often used instead as the main means of protection against imitation (see Barton,
1993; Samuelson, 1993). We readily admit this to be the major empirical shortcoming
of our analysis, and plead only that no-one has yet found a satisfactory, accessible and
systematic measure of competencies in software technology that we could uses. And
as we shall see in section 2 below, we have nonetheless been able to identify the
growing importance of competencies in information technology.
The third limitation to our analysis is that we do not assess how differences in the rate and
direction of technological accumulation affect firms' economic and competitive performance.
Suffice to say that a growing number of studies confirm the competitive importance of
technological competencies at the level of the firmg, which should in principle heighten
interest in studies like ours that attempt to describe and explain how they are acquired.
1.5 Structure
We shall now describe the key characteristics of large firms' technological competencies that
emerge from our analysis: diversity in section 2, differentiation and stability in section 3, and
the influence of sector, country and management in section 4. In section 5, we draw
conclusions for practice and for theory.
2 TECHNOLOGICAL DIVERSITY:
THE PREVALENCE OF THE "MULTI-TECHNOLOGY" FIRM
2.1 The Extent of Technological Diversity
The most striking feature of the technological competencies of large firms is the diversity of
technological fields in which they are active. This is shown most simply in Table 1, which
gives the distribution of US patenting of our large firms, in each of the 16 principal product
groups, across four major technological families: chemical, electrical+lectronic, nonelectrical machinery and transport, as well as a residual category labelled 'other'lo. Firms
have substantial technological competencies outside what would appear to be their core areas.
Thus, both electrical and chemical firms have about two-thirds of their competencies in their
obvious core areas, but each has 15% or more in non-electrical machinery: and automobile
Recent research by Jacobsson and Oskarsson (1994) uses very interesting data on the technical field of
specialisation of Swedish engineers working in Swedish firms. Unfortunately, this method cannot easily be
reproduced in other countries, because of lack of data.
See, for example, Cantwell (1989). Franko (1989), Geroski et al., (1993), Oskarsson (1993).
lo The method for distributing firms' technological activities amongst four technological families is described
more fully in Patel and Pavitt (1994b). Briefly stated we re-classified the US Patent Classes and sub-classes
into 34 technical fields, and 91 sub-fields. On the basis of the 91 sub-fields, we re-combined patenting into the
four technological families shown in Table 1. The "Other" category includes traditional manufacturing (e.g.
textiles) and non-manufacturing (e.g. construction, medicine, agriculture).
firms have less than a third of their competencies in transport technologies, but more than
45% in non-electrical machinery. Only firms principally in pharmaceuticals have less than
10% on average of their technological competencies in non-electrical machinery.
Table 1. The Distribution of Large Firms' Technological Activities in Five Broad
Technological Fields, according to their Principal Product Group: 1981-90.
Principal Product Group
(PPG)
Chemicals
Pharmaceuticals
Mining & Petroleum
Textiles etc.
Rubber & Plastics
Paper & Wood
Food
Drink & Tobacco
Building Materials
Metals
Machinery
Electrical
Computers
Instruments
Motor Vehicles
Aircraft
All 440 Large Firms
Percentage share of the PPG's patents in technology field
NonElectrical
Chemical Machinery Electrical Transport Other
16.9
8.9
0.6
2.6
7 1.O
2.1
0.0
9.7
80.2
8.O
57.1
34.2
6.7
0.9
1.1
52.9
3 1.7
9.5
0.6
5.3
43.2
29.3
4.7
20.1
2.7
25.4
47.1
12.4
0.4
14.6
70.6
21.9
3.0
0.1
4.3
40.8
50.3
4.6
0.3
3.9
10.0
0.9
7.3
30.5
5 1.3
26.8
54.9
13.9
2.1
2.2
7.6
64.9
13.9
10.2
3.3
7.6
21.2
67.0
1.3
2.8
5.2
16.3
77.3
0.2
1.0
14.3
18.3
64.2
0.1
3.0
3.8
44.8
20.7
28.8
1.9
8.1
48.5
31.2
8.3
3.9
28.8
27.9
35.7
4.4
3.1
Total
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
100.0
Source: Calculated from data supplied to SPRU by the US Patent and Trademark Office.
Another measure of technological diversity is the number of technical fields - out of the total
of 34 used in our analysisl1- in which our firms have been granted a patent and are therefore
technically competent. Table 2 confirms this diversity: only 4% of our firms were active
sometime in the 1980s in 10 or fewer of these technical fields, whist 52% were active in
between 10 and 20, and 44% in more than 20 - hence the term "multi-technology" firm (See
Archibugi, 1988; and Granstrand and Sjolander, 1990)'2.
'
See Table 5 for the name of each of the technical fields.
The distribution of our firms amongst the different degrees of technological diversity shown in Table 2 is
sensitive to the measure of technological competence chosen. Thus, when it is increased from one to ten patents
in the 1980s. the proportion of firms active in more than 10 technical fields declines from more than 95% to just
over 30%. However, as we shall show in section 3, apparently low-level technological activity is an important
and permanent feature of firm-specific technological competencies. And other measures confirm large firms'
technological diversity. For example, 90% of total technological activity is concentrated in five or fewer
technical fields in 14 % of our firms, whilst 64% reach this threshold at between 6 and 10 fields, and 20% with
more than 10 fields.
l2
Table 2. Number of Technical Fields (out of 34) in which Firms have one Patent or
more in 1981-90: Percentage Distribution.
Product Group
Number of Less than
Greater
Greater
firms or equal to than 10 but than 20 but
10 less than or less than or
equal to 20 equal to 30
39.4
66
4.5
50.0
Chemicals
56.0
32.0
Pharmaceuticals
25
12.0
Mining & Petroleum
31
0.0
48.4
38.7
10
10.0
80.0
10.0
Textiles etc.
9
0.0
77.8
Rubber & Plastics
22.2
83.3
18
5.6
11.1
Paper & Wood
14
42.9
42.9
14.3
Food
100.0
0.0
8
0.0
Drink & Tobacco
56.3
43.8
Building Materials
16
0.0
38
0.0
57.9
42.1
Metals
58
1.7
67.2
31.0
Machinery
37.5
56
0.0
48.2
Electrical
17
11.8
58.8
29.4
Computers
21
4.8
38.1
57.1
Instruments
48.6
35
2.9
48.6
Motor Vehicles
18
0.0
22.2
77.8
Aircraft
All Sectors
440
4.3
52.0
40.0
Greater
than 30
Total
6.1
0.0
12.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
14.3
0.0
0.0
0.0
0.0
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
100.00
3.6
100.00
Source: Calculated from data supplied to SPRU by the US Patent and Trademark Office.
2.2 The Determinants of Technological Diversity
We have suggested elsewhere (Pavitt, 1984; Pate1 and Pavitt, 1992) that two factors influence
the degree of diversity of large firms' technological activities:
1. Firm Size will be positively associated with technologically diversity, both as a
consequence of successful product diversification in science-based technologies
(chemicals and electrical~lectronics),and as an incentive to the application of
production-based technologies in order to exploit economies of scale. We would
therefore expect a positive association between firm size and technological diversity,
measured as the number of technological fields in which the firm is active.
2. Technology Intensity will also influence a firm's technological diversity. Increased
intensity - measured as patenting per unit sales - will be positively associated with the
of fields of competence, reflecting the results of more energetic technological search.
3. Home country characteristics are also said to influence firms' degree of technological
diversity. For example, it is argued that the competitive and institutional framework
for Japanese firms leads them towards greater technological diversity than in other
countries (see, for example, Kodama, 1986; and Oskarsson 1993).
4. Finally, it can be argued that industry characteristics influence the number of fields in
which the firm is active through the range of competencies required to develop and
produce a given class of products.
In Table 3, we present the results of our regressions testing the above explanations. The
dependent variable is each firm's number of active fields of competence (out of a total of 34)
in the 1980s; the independent variables are each firm's sales, patent intensity, country of
origin and industry. The results show that the coefficients on size and technology intensity
have the expected sign and are significant at the 5% level. On the other hand firms' countries
of origin have no significant effects on the diversity of technological competencies, since the
country dummy variables are not significant at the 5% level. Industries (i.e. product groups
made) do matter, with food firms showing the least technological diversity and aircraft firms
the most.
Table 3. Determinants of Technological Diversification
Dependent Variable: Number of Technical Fields (out of 34) of Patenting (81-90)
Coeff.
Std Error
Coeff.
Std Error
Constant
16.1l*
0.34
15.16*
1.33
0.25*
0.02
0.25*
0.02
Sales (1988)
Patent Intensity (1988)
0.02*
0.00
0.02*
0.00
Dummy Japan
-0.95
0.55
Dummy USA
0.45
0.48
2.98*
1.39
Dummy Chemicals
Dummy Pharmaceuticals
-1.46
1.54
Dummy Mining & Petroleum
1.98
1.5 1
0.2 1
1.87
Dummy Rubber & Plastics
-1.89
1.61
Dummy Paper & Wood
1.70
Dummy Food
-3.9 1*
Dummy Drink & Tobacco
-1.22
1.94
Dummy Building Materials
1.71
1.65
Dummy Metals
1.84
1.45
Dummy Machinery
1.06
1.40
Dummy Electrical
2.62
1.42
Dummy Computers
-1.26
1.63
Dummy Instruments
-1.61
1.68
Dummy Motor Vehicles
-0.56
1.49
Dummy Aircraft
4.37*
1.62
* Indicates that the coefficient is significantly different from zero at the 5% level.
8
2.3 Changing Technological Diversity over Time
Not only are large firms technologically diverse, but their diversity has been changing over
time. This is confirmed in Table 4, which shows the numbers of firms (from Europe, Japan
and the USA) whose technological diversity increased, decreased and remained stable over
this period. It emerges clearly that firms differ markedly according to their country of origin,
with most Japanese firms increasing the technological diversity of their patenting activities,
and a majority of European firms doing likewise, whilst most US firms decreased the
diversity of their patenting. At the sectoral level, technological diversity increased in US
firms in pharmaceuticals, computers and drink and tobacco, and in European firms in
chemical and machinery related sectors.
Table 4. Changes in Firms' Technological Diversity by Product Group and Region:
1969-74 to 1985-90.
Number of Firnzs
USA
Dec Stab Inc
Europe
Dec Stab Inc
Japan
Dec Stab Inc
Total
Dec Stab Inc
Chemicals
Pharmaceuticals
Mining & Petroleum
Textiles etc.
Rubber & Plastics
Paper & Wood
Food
Drink & Tobacco
Building Materials
Metals
Machinery
Electrical
Computers
Instruments
Motor Vehicles
Aircraft
23
3
10
3
5
7
8
2
7
8
20
16
5
6
6
10
0
3
1
2
0
1
2
2
1
3
2
4
0
2
0
0
3
8
5
1
0
5
2
3
1
4
8
7
5
3
4
2
3
2
2
1
3
1
0
1
3
7
7
5
1
0
5
2
2
0
1
1
0
0
0
0
0
3
1
3
1
1
1
0
10
5
8
1
0
2
1
2
1
9
12
4
2
0
9
5
0
0
0
0
0
0
0
0
0
0
1
0
0
2
0
0
1
0
1
0
0
0
1
0
0
0
0
0
0
0
0
0
23
4
3
4
2
1
2
1
4
10
11
17
2
5
12
0
26
5
12
4
8
8
8
3
10
15
28
21
6
8
11
12
All Product Groups
139
23
61
43
14
72
3
3 101
185
3
3
3
3
0
1
3
2
1
6
3
7
1
3
1
0
36
17
16
6
2
8
5
6
6
23
31
28
9
8
25
7
40 234
Dec: Firms where there has been a decrease in the number of technical fields (out of 34) of activity.
Stab: Firms where there has been no change in the number of technical fields (out of 34) of activity.
Inc: Firms where there has been an increase in the number of technical fields (out of 34) of activity.
The meaning of these trends is ambiguous. It is tempting to conclude that the declining
technological diversity of US firms reflects their declining technological competitiveness,
compared to firms from Japan and Europe13. However, the data for US firms reflect domestic
patenting, the scope of which is sensitive to its cost; whilst the data for European and
Japanese firms also reflect international patenting, the scope of which reflects international
l 3 Since the late 1960s, business-funded R & D has increased more rapidly in Japan than in Europe, and more
rapidly in Europe than in the USA. See Patel and Pavitt, 1994a.
technological competitiveness and business strategy. The trends could therefore simply
reflect increases in the cost of US patenting (influencing US firms), and the processes of
technological catch-up (influencing European and Japanese firms). Suffice to say that, by the
1980s, our US firms were in aggregate still slightly more diversified (according to the same
measure as in Table 2) than the European and Japanese firms.
Table 5. Changes in the Number of firms that are Active in 34 Technical Fields, by
Region: 1969-74 to 1985-90.
Sorted bv total change
Calculators & Computers, etc.
Drugs & Bioengineering
Materials (inc glass & ceramics)
Plastic & rubber products
General Electrical Ind. Apparatus
Instruments & controls
Metallurgical & Metal Treatment proc.
Dentistry & Surgery
Miscellaneous metal products
Other - (Ammunitions & weapons, etc.)
Image & sound equipment
Chemical Processes
Mining & wells: mach. & proc.
Hydrocarbons, mineral oils, fuels etc.
General Non-electrical Ind. Equip.
Agricultural Chemicals
Semiconductors
Photography & photocopy
App. for chemicals, food, glass etc.
Assembling & material handling app.
Road vehicles & engines
Electrical devices & systems
Organic Chemicals
Non-electrical specialized ind. equip.
Power Plants
Inorganic Chemicals
Aircraft
Metallurgical & metal working equip.
Telecommunications
Bleaching Dyeing & Disinfecting
Other transport equip. (exc. aircraft)
Food & Tobacco (proc. & prod.)
Induced Nuclear Reactions
Textile, clothing, leather, wood products
WE denotes European Firms.
WE
74
1985-90
JP US Total
69 142 285
Change Since 1969-74
WE
JP US Total
14
34
22
70
Stronger conclusions can be reached about the technical fields into (and out of) which firms
are moving over time. Table 5 shows the total number of large firms that have been active in
each of our 34 technical fields in 1985-90 and the changes therein since 1969-74. It thereby
compares the degree of pervasiveness of technological competencies in different fields, and
how this has changed over time. The technological fields are sorted according to the last
column, namely, the change in the number of active firms between the two periods. It
emerges that:
for firms from Japan, Europe and the USA, the most pervasive competencies are the
same: instrumentation and control, production machinery and chemical processes, in
all of which the overwhelming majority of our firms was technologically active;
the least pervasive competencies were in nuclear energy, aircraft and textiles;
over time, the sectors in which the number of firms with competencies increased most
rapidly were computing, drugs and bio-engineering, and materials;
the patterns and trends were similar in all three regions, except for a particularly sharp
decline in US firms with competencies in image and sound, and in photography and
photocopy.
2.4 Some Implications of "Multi-technology" Firms
Our results are consistent with the conclusions of research by Ove Granstrand and his
colleagues at Chalmers in Swedenl4. In particular, large firms and the products they make
depend on many fields of technological competence, the number of which is changing over
time with the widening range of technological opportunities emerging from improvement in
computing and other science-based technologies.
In order to assimilate this range of
emerging technologies, large firms simultaneously increase their internal competencies, form
alliances with external sources, and increase their overall R & D expenditures.
At the same time, the striking technological diversity of our large firms casts some doubt on
the feasibility of a "focused" technological - as distinct from product market
-
strategy
(Porter, 1985), given that the products that they make are multi-technology (see Freeman,
1982). Similarly, the notion of "core competencies" (see Prahalad and Hamel, 1990) in
technological strategy is not entirely clear, when large firms are typically active so many
technical fields.
l 4 See, in particular, Granstrand and Sjolander, 1990; Granstrand et al., 1992; Jacobsson and Oskarsson (1994);
Oskarsson, 1993.
In this context, it is worth noting that business practitioners often have a more elaborate
classification of firms' technological competencies. According to those in large firms that are
members of the European Industrial Research Management Association (EIFWA):
"In order to consider explicitly the technological resources needed to implement a
strategic plan, it is essential to know the precise technological position of the company (or
the business unit), in relation to that of its major competitors.. ....This can .......be
considered in terms of three types of technologies which show differing potential for
competitive impact.
+ Basic technologies - Widely available; low risk low reward.
+ Key technologies - Proprietary; essential to maintain in-house; medium risk,
medium reward.
+ Pacing technologies - These can produce a breakthrough for the company: normally
achieved by in-house effort over a long time; high risk, high reward."
(EIFWA, 1986, p. 19)
3 FROM FIRM-SPECIFIC COMPETENCIES TO PROFILES
In this context, we shall now show that large firms have profiles of competencies, with levels
of commitment and advantage that vary amongst technological fields. We shall also show
that these profiles are highly stable, differentiated and strongly related to the product base.
3.1 Defining and Measuring Firms' Technological Profiles
Our definition of a firm's technological profile reflects both the experience of practitioners
(see quote from EIRMA above) and the nature of our data base, as well as earlier
contributions to the subject. We distinguish two interrelated dimensions of technological
competencies.
1. Core vs. Niche Competence reflects the relative importance of the field in the firm's
total portfolio. It is measured as the share of the firm's patenting in each of our 34
technical fields (PS). Relatively high shares will measure what we call a core
competence, and relatively low shares a niche competence.
2. Distinctive vs. Background Competence reflects the degree to which the firm has an
advantage in the field compared to other firms. It is measured as the firm's share of
total patenting in the field, divided by the firm's aggregate share in all fields.
Elsewhere, we have called this the Revealed Technology Advantage (RTA) of the firm
in each field. A high RTA will measure what we call a distinctive competence, and a
low RTA what we call a background competence.
We represent the full classification in Figure 1 below, showing that firms can have four
categories of technological competence (in addition to having no competence of any kind in
some fields). The following properties are of particular importance.
Some categories are more important than others - in particular, core-distinctive is more
important than niche-background.
Given their definition, the measures along the two axes are correlated, and the
correlation would be perfect, if there were an equal volume of total patenting in all 34
technical fields.
However, there are technical fields with relatively low levels of total patenting activity,
where firms may develop a niche-distinctive competence.
There are also technical fields with relatively high levels of patenting activity, where
many firms have a core-background competence that is very similar to what European
practitioners call basic technologies1? Our own earlier analysis shows that large firms
find it necessary to maintain some in-house competence in basic (background)
technologies, that are often related to production techniques and located in fields of
mechanical, chemical and instrumentation engineering (Pate1 and Pavitt, 1994b. See
also Table 5 above).
The positioning of the axes on Figure 1 is (inevitably) arbitrary. We have placed the
line between core and niche at the share that would allow equal distribution across all
fields: 100134 = 3%.
We have defined the difference between distinctive and
background (and the cut-off for the latter) more pragmatically, after examining the
profiles of a number of firms.
Core
[PS > 3%]
Core-Background
Core-Distinctive
Background
[2.0 > RTA > 0.51
Distinctive
[RTA > 2.01
Niche-Background
Niche-Distinctive
I Niche
[PS < 3%]
Figure 1: A Classification for Firms' Technological Profiles
I s Available not only within large firms, but also in smaller, specialised firms to whom large firms sometimes
"spin-off their innovations (Rosenberg, 1976; von Hippel, 1988).
In Figure 2 below, we reproduce the technological profiles of three large (and well-known)
firms, from the chemical electrical and automobiles industries'b. A number of features of
firms' profiles emerge from Figure 2.
A relatively large number of technological fields combine to define each firm's
technological profile: 11 in chemicals, 18 in electrical, and 20 in automobiles.
In all three firms, these sectors account for more than 90% of the firms' patenting.
The core distinctive competencies are very different:
- chemicals: organic and agricultural chemicals, pharmaceuticals and photography;
- electrical: computers, semiconductors, and image and sound;
-
automobiles: vehicles, engines and other transport.
All three firms have at least one niche-distinctive competence;
The chemical firm has just one core-background competence (chemical processes)
accounting for 7% of all its patenting.
The electrical and automobile firms are very different, with respectively 8 and 10 core
background competencies, accounting for 47% of all patenting in the electrical firm,
and 64% in the automobile firm. In both cases, instrumentation accounted for about
15% of all patenting activities.
Q r g Che
Orugs
Photo
C h e m Proc
Wgri Chem
materials
C h e m App.
P l a s ic
-leaching B Dyeing
Wssemb Equip
Mnorg Chem
Figure 2a. Technological Profile of a Chemical Company
l6
Since we are (amongst other things) interested in illustrating differences amongst firms from different
industries, the RTAs are calculated on the basis of patenting by firms from all sectors. For competitor analysis,
they should probably be calculated on the basis of competitor firms only.
.Instrur ents
.1ma~e%GR8'~~~
IElectr Equip
Gemicond
=Nan El Mach qelecoms
Chem ProG~lectrDevi
r
.Metal Work Equip
.Chem
mPhotogr
Wetallurg
Handling Equip
Wuclear
.Power Plants
Figure 2b. Technological Profile of an Electrical Company
Web. Eng.
\ m€&?cEquip
C a m Proc
Computers
.Oth Trans
.MaterlalS
8Semicond
=In Org Che
$r%\dlurg
.Aircrafl
=Tex.Cloth.Wood
-Power Plants
Figure 2c. Technological Profile of an Automobile Company
On the basis of these examples, we may provisionally conclude that the classification of
technological profiles proposed in Figure 1 has three potential strengths.
1. It encompasses the wide variety of technological competencies accumulated within
large firms.
2. It distinguishes the differing contributions of each field of competence.
3. It highlights the importance of the core background competencies that are often
neglected or ignored in conventional analysis.
We shall now show that large firm's technological profiles have two other characteristics:
they are both highly stable and highly differentiated.
3.2 The Strong Stability of Firms' Technological Profiles and Directions of Search
For nearly all our firms, these technological profiles are remarkably stable over time. For
each firm, we correlated both the patent shares (PS) and the RTAs for the periods 1969-74
and 1985-90. Table 6 shows that according to both measures, the overwhelming majority
(more than 90%) of firms have profiles of technological competence that are statistically
similar between 1969-74 and 1985-90, at the 1% level of significanceI7. Large firms clearly
do not shift around rapidly in their fields of technological competencelg.
Table 6. Stability of Technological Profiles Across 34 Technical Fields: 1969-74 to
1985-90.
Revealed Technology
Patent Shares
Advantage
No. of
Not Sig Not Sig Not Sig
Not Sig Sig at Sig at
Firms
at 5%
5%
1%
at 5% at 5% at 1%
1 Chemicals
65
1
1
63
5
7
53
2 Pharmaceuticals
25
2
3
20
0
0
25
3 Mining & Petroleum
31
7
17
5
5
21
7
4 Textiles etc.
13
4
6
3
5
6
2
5 Rubber & Plastics
10
0
0
10
1
1
8
6 Paper & Wood
17
1
3
13
4
4
9
7 Food
16
0
1
15
1
2
13
8 Drink & Tobacco
11
0
1
10
0
2
9
9 Building Materials
17
0
0
17
0
0
17
10 Metals
44
4
6
34
5
7
32
11 Machinery
63
2
5
56
5
10
48
12 Electrical
56
4
5
47
5
8
43
13 Computers
16
0
0
16
0
1
15
14 Instruments
19
0
0
19
2
3
14
15 Motor Vehicles
37
0
0
37
2
2
33
16 Aircraft
19
0
19
1
1
17
0
All Sectors
459
25
38
396
41
59
359
No systematic differences in stability can be detected between firms in different sectors and countries.
Given our method of compiling data of firm-level patenting, we cannot measure any changes in our firms'
technological profiles resulting from acquisitions and divestments. On the basis of data for large Swedish firms,
Oskarsson (1993) has concluded that acquisitions and divestments have had little influence on the shape of their
technological profiles.
l7
This stability over time in firms' technological profiles is defined by relatively broad
technological fields, and does not reflect the more detailed processes of search that firms
undertake. For this reason, we have identified in US patenting activities the 1,000 (out of a
total of around 100,000) technological sub-classes of the highest technological opportunity, as
measured by their absolute increase in patenting from the 1960s to the late 1980s. In
aggregate, their share increased steeply from 3 to 18% of total US patenting. A relatively
high proportion of fast growing fields (FGFs) are to be found in electronics and chemical
technol~gies'~.
In Table 7, we show that firms are in fact heavily constrained by their prior competencies in
the directions in which they accumulate competencies in these fast-growing fields. Their
shares of total fast-growing patenting in 1985-90 within the five broad fields of technology
used in Table 1 - chemicals, mechanical, electrical~lectronic,transport and "other" - are
strongly and positively correlated with their prior shares of total patenting in these same fields
over the period 1969-84. In other words, firms' capacities to exploit fields of high
technological opportunity are strongly constrained by their prior competencies.
Table 7. Correlations of Past (1969-84) Shares of Total Patenting on Shares of
Patenting in Fast-Growing Areas in 1985-90.
Share of Total Chem 69-84
Share of Total Mech 69-84
Share of Total Elec 69-84
Share of Total Trans 69-84
Share of Total Othe 69-84
Shares of Patenting in Fast-Growing Areas in 85-90
Chemicals Mechanical Electrical Transport
-0.6 1*
-0.26"
-0.4 1*
0.9 1*
0.68"
-0. lo*
-0.4 1*
0.14"
-0.12"
0.87"
-0.58"
-0.17"
0.18"
-0.13"
-0.34"
0.85"
0.06
-0.12"
-0.18"
-0.07
Other
0.00
0.09"
-0.17"
-0.04
0.55"
* Denotes a coefficient significantly different from zero at the 5% level.
3.3 The Differentiation of Industries' Technological Competencies
In addition to being very stable, our data also show that large firms' technological
competencies are highly differentiated. To begin with, average patent shares and RTAs for
each of our sixteen industries (i.e. aggregate data based on our firms) are in general very
different. For patent shares, 23% of the cross-industry correlations are positive and
significant20; and for RTAs the share is reduced to 5%. In both cases, there are essentially
three clusters:
l 9 For this reason, we find significant correlations between firms' share of total patenting in fast-growing fields,
on the one hand, and their R & D intensity and share of total patenting in science based technologies, on the
other.
O'
At the 5% level.
the chemical and chemical-related industries (the first eight listed in Table 1);
machinery and vehicles;
electrical and computers.
There is also one significantly negative correlation that is important: between the RTAs of
firms in chemicals and in electrical products. Although both are often lumped together as
"high technology" or "science- based" firms, they are clearly based on very different mixes of
technological competence.
The statistical similarities and (above all) differences described above reflect similarities and
differences in core and distinctive competencies amongst firms in different sectors. These are
set out systematically in Table 8 which describes the contribution of competencies in our 34
technical fields to firms in each of the 16 sectors according to the four-fold classification
shown in Figure 1.
From Table 8 it emerges that technical fields vary greatly in the nature and extent of their
contributions to firm-specific competencies:
organic chemicals and materials are core distinctive competencies in five industries;
drugs, non-electrical machinery, and image and sound in three each; instruments (in
spite of its overall importance) in only one; and five fields in none at all;
as can be anticipated from Table 5, core background competencies are located mainly
in chemical processes, machinery, instrumentation, and organic chemicals;
niche distinctive competencies are restricted to relatively few fields such as plastics,
dyestuffs, nuclear energy and power plant;
the most prevalent of niche-background technologies are assembly and materials
handling, plastic and rubber, and metallurgical processes;
in spite of its spread amongst an increasing number of companies, computer
competence is so far identifiable beyond the usual "high-tech" industries only in
machinery and vehicles.
It also emerges from Table 8 that profiles of competencies vary greatly amongst firms in the
different sectors:
the number of technical fields involved varies from 7 in pharmaceuticals to 24 in
aircraft;
in only four sectors (chemicals, pharmaceuticals, petroleum and mining and electrical
products) do the number of core distinctive technological fields outnumber the number
of core background fields;
a
in at least six sectors, core background fields account for more than 50% of all
technological competencies.
3.4 Do Firms' Technological Profiles match Product Groups?
One drawback in our analysis so far is that it neglects the possibility of diversity in the profile
of technological competencies of firms within each industrial sector. For this reason, we
summarise in Tables 9 and 10 of our systematic examination of the similarities and
differences in profiles of technological competencies individually for all our large firms.
Each table shows the percentage of firms' technological profiles, for the period 1981-90, that
are similar (that is - positively correlated at the 5 % level) to firms inside the same product
group, and to those in the other product groups; Table 9 does this for patent shares, and Table
10 for Revealed Technology Advantage. The main patterns that emerge are as follows.
a
Firms have significantly different profiles of technological competence to most others:
19 % are similar in patent shares (i.e. core competencies), and 11% in RTAs (i.e.
distinctive competencies).
More generally, firms are more similar (or less dissimilar) to each other in their core
than their distinctive competencies.
In all sectors, firms have a higher probability of finding others with similar
technological profiles within their sector than outside: from three times as high for
machinery firms (according to RTAs), to more than fourteen times as high for
pharmaceutical firms.
The frequency of technological proximity between firms in different sectors is not
evenly spread or random, but reveals distinct groupings, many of which have been
anticipated earlier in Table 8: in particular, those with competencies in organic
chemicals, in electronics, and in production machinery.
These sectoral similarities and differences amongst firms in the sources and directions
of technological accumulation are broadly consistent with a sectoral taxonomy of
technical change proposed earlier by one of us (Pavitt, 1984):
-
two distinct science-based sectors centred on organic chemistry (chemicals,
pharmaceuticals, petro-chemicals), and on physics-based technology (electrical,
computers);
-
machinery suppliers with areas of specialisation influenced by major users;
-
a range of scale intensive sectors with production technologies dependent on
improvements in chemical processes, instrumentation and production machinery.
Table 9. Correlations of Firms' Shares across 34 Technical Fields, by Principal Product Group: 1981-90.
Percentage of the total that are Positive and Significant at 5% level.
Own
All Other
PPG
PPG's
Phar Mini Text Rubb Pape Food Drin Buil Meta Mach
Chemicals
78.6
19.1
60.3 61.8 53.6 49.0 17.8 39.0
9.8 24.9 19.1
4.5
Pharmaceuticals
Mining & Petroleum
Textiles etc.
Rubber & Plastics
Paper & Wood
Food
Drink & Tobacco
Building Materials
Metals
h.' Machinery
Electrical
Computers
Instruments
Motor Vehicles
Aircraft
Elec Comp
1.7
0.0
Inst Moto
8.7
2.4
Airc
1.9
w
w
Table 10. Correlations of Firms' RTA's across 34 Technical Fields, by Principal Product Group: 1981-90.
Percentage of the total that are Positive and Significant at 5% level.
Own
All Other
PPG
PPG's
Phar Mini Text Rubb Pape Food Drin Buil Meta Mach Elec Comp Inst
Moto Airc
Chemicals
48.7
9.6
26.1 25.2 33.6 26.6 12.0
6.4
2.7 19.1 14.5
3.8
1.3
0.0
2.5
0.8
0.9
Pharmaceuticals
86.7
6.2
3.1 16.8
0.9
3.8 20.6 13.0
0.5
0.0
0.5
0.1
0.0
4.2
0.0
0.0
Mining & Petroleum
72.9
7.4
4.2
7.5
9.7
0.2
0.0
4.6 14.2
3.6
0.6
0.0
1.1
1.8
2.9
Textiles etc.
46.7
12.4
33.3 23.3 10.0 10.0 38.1 10.8
4.5
4.8
0.0
3.3
0.3
0.0
Rubber & Plastics
86.1
9.0
17.3
0.0 11.1 18.1
1.5
9.0
2.0
0.0
0.5
0.0
9.3
Paper & Wood
37.3
7.9
11.9 22.2 29.9
1.6
9.2
2.8
0.7
6.3
1.4
0.0
Food
100.0
5.2
87.5
6.3
0.8
1.8
0.1
0.0
0.0
0.0
0.0
Drink & Tobacco
82.1
6.9
10.2
2.0
5.8
1.1
0.0
0.0
1.1
0.0
Building Materials
52.5
9.6
10.5 10.0
3.9
0.0
6.0
2.1
1.0
Metals
77.8
7.2
11.4
5.1
0.0
0.9
3.8
3.2
Machinery
21.3
7.4
5.8
2.0
3.1 27.1
7.6
Electrical
45.5
5.7
47.4 15.3
3.6
5.5
Computers
99.3
7.6
16.5
0.0
3.9
Instruments
38.1
4.9
1.8
3.7
Motor Vehicles
75.6
5.7
8.6
Aircraft
76.5
3.7
4 COMPETENCIES AND MANAGERIAL CHOICE:
THE EFFECTS OF PRODUCT MIX AND HOME COUNTRY
It is already clear from the above analysis that managerial choice is constrained by firm's size
and product mix. In particular, we have shown that:
large firms are generally technologically diversified, and slowly changing over time ,
as the range of technological opportunities increases;
however, each firm's profile of technological competencies remains very stable, and is
strongly constrained by the products it makes;
similarly, each firm's direction of technological search (and accumulation of
competence) is strongly constrained by its prior competencies.
In other words, a firm's existing product mix and associated competencies strongly constrain
the directions in which it seeks to exploit technological opportunities and acquire competence.
We shall now extend these analyses to explore the determinants, not of the direction, but of
the rate of the firm's technological search activities. We suggest that three factors will
influence the rate of search.
1. The firm's home country will influence its rate of technological accumulation through
the nationally-based supply and demand-side inducement mechanisms described by
Porter (1990). These are likely to remain strong since, globalisation of markets and
other things notwithstanding, large firms continue to perform the overwhelming
proportion of their R & D activities (-90%) in their home countries (Pate1 and Pavitt,
1991; Patel, 1994).
2. The firm's sector of activity will influence its rate of technological accumulation.
Given that the firm's competencies and directions of search are determined in large
part by what it produces, and that technological opportunities are unequal across fields,
firms will have varying capacities to exploit opportunities, and thereby varying rates of
accumulation (Malerba, 1992).
3. Firm-specific factors will also influence the rate of technological accumulation. Given
uncertainties, different managements will make different bets. Also, the professional
background of managers, and their associated "rules of thumb" and professional
loyalties, may influence the propensity to encourage technological accumulation21.
In Table 11, we present the results of our analysis of the effects of home country conditions
and of product mix (both measured through the appropriate aggregate indicators from our
2 1 See, for example, Scherer and Huh (1992) and Bosworth and Wilson (1992), who have shown that, in the
USA and the UK, the level of firms' allocation of resources to technological activities is positively associated
with the presence of graduate scientists and engineers in top management.
large firm database) on various measures of the rate of accumulation of technological
competencies in firms. From this it emerges that:
both home country and product mix have a statistically significant influence on the rate
of technological accumulation, whether measured in terms of patent per unit of sales,
growth in patent share, or share of total patenting in fast-growing fields;
the unexplained variance amongst firms nonetheless remains considerable - 56-80%
of the total, which suggests that company-specific factors - and particularly those
influencing the volume of resources allocated to technological accumulation - remain
important22.
Table 11. Factors Influencing Firms' Rate of Technological Accumulation.
Dependent Variable
Constant
Industry Average
Country Average
Patent Intensity: Change in Patent Share: Share of Patents in F-G:
1985-90
1985-90
1969-74 to 1985-90
Coeff. Std Error
Coeff. Std Error
Coeff. Std Error
0.83
0.52
-6.7
1*
1.69
-32.08*
10.90
1.88*
0.42
0.82*
0.05
1.03*
0.07
0.68*
0.07
0.38*
0.07
0.62*
0.16
* Denotes Coefficient Significantly Different from zero at the 5% level.
5 CONCLUSIONS
The substantive findings of our paper are surnrnarised at its beginning. We shall now explore
their implications for policy, theory and the agenda for future research.
5.1 For Corporate Policy
Our results confirm and sometimes clarify some prescriptions on technology strategy in large
firms, whilst casting doubt on others.
The Strong Constraints on Feasible Choice. Perhaps most important, we have identified a
number of important constraints on the strategies for technological competence-building in
large firms.
22 Since all three dependent variables are based on patenting, part of the unexplained variance may reflect interfirm differences in the propensity to patent the results of R & D and related technological activities. However,
this is less likely to operate in shares of total patenting in fast-growing fields, where more than 55% of inter-firm
variance still remains to be explained.
Their technological strategies can only rarely be "focused", since the products they
develop and make require the integration of knowledge from a range of technological
fields (see also Freeman, 1982).
Their competence to exploit specific technological opportunities is highly
differentiated, and heavily dependent on past competencies accumulated through
making specific classes of products.
Their capacity to modify their profiles of technological competence is limited, and
takes a long time (see also Rosenbloom and Cusumano, 1987).
In addition to these constraints on the directions of technological accumulation, both
home-country and industry characteristics have a significant influence on the firm's
rate of competence accumulation, although the unexplained variance suggests
considerable scope for managerial discretion.
The Nature of Competencies. We have also shown that the simple concept of "core
competence" cannot encompass the range and variety of technology profiles found in large
firms. Our four-fold classification shows that large firms also accumulate background - and
manly production-related - competencies, the importance of which can sometimes outweigh
distinctive competencies. This variety of inputs into large firms' competencies has other
important consequences that we shall now explore.
Technology Fusion. The growing spread of firms' technological competencies in computing,
biotech and materials must be seen in the context of the widespread and continuing
importance of competencies in mechanical, chemical and instrumentation engineering. As
Kodama (1986) has pointed out, effective innovation often requires the combined
development of well-established and new technological fields, and takes considerable time,
which is why firms' technological profiles change only slowly, and why the implementation
of corporate technology strategy requires the effective integration of firm-specific-knowledge
across a range of fields (see, for example, Henderson and Cockburn, 1993).
Competence-Destroying Innovations? It is also why "competence-destroying" innovations
are less likely in large firms with diversified competencies and R & D programmes (Cooper
and Schendel, 1976; Tushman and Anderson, 1987; Utterback and Suarez,1993). Although
radical breakthroughs may destroy one part of such a firm's competence, it is unlikely to
destroy them all. This can be seen in the new biotechnology, where - in spite of a slow start established chemical and pharmaceutical firms have succeeded in combining the radical
breakthroughs with their established fields of competence (Arora and Gambardella, 1992;
Galimberti, 1993).
"Technological Leap-frogging"? Similarly, the multi-technology nature of many products
reduces possibilities for "technological leap-frogging" by developing countries. Perez and
Soete (1988) have suggested that radical new technologies open opportunities for late-comer
countries and companies to by-pass (or "leapfrog") earlier paths of technological
accumulation, and to enter new product areas with greater dynamism than more advanced
countries and companies more firmly embedded in older paths of technological development.
However, when new technologies must be combined or "fused" with older ones, the earlier
paths of technological accumulation cannot be avoided. This remains the case in many
sectors of central importance to economic development, including the electronics sector in
East Asia, which depends heavily on technological competence in production engineering
(Bell, 1984; Hobday, 1993).
External Strategic Alliances as Complements to In-house Competence-Building. Finally,
we should note that the technological fields where firms have been acquiring an in-house
capability most vigorously since the early 1970s - computers, biotechnology and
pharmaceuticals, and materials (see Table 5) - are also those where firms have increased most
vigorously their external alliances for technological exchanges and joint developments (see
Mowery, 1988; Hagedoorn and Schakenraad, 1992). This suggests - as Granstrand and his
colleagues have noted in micro-studies (1992) - that external alliances are complements to
internal learning, and not substitutes.
5.2 For Theory
Complexity and Path-dependency. The variety, stability and differentiation of large firms'
technological competencies, their close links to the products that they make, and the localised
nature of their technological search, all confirm the importance of technological (and other)
complexity in:
constraining firms' processes of technological search (see, for example, David, 1975);
allowing firms to differentiate themselves to explore and master different zones of
technological complexity;
explaining why the assimilation of radical new technology takes a long time.
Technological Relatedness rather than Economies of Scope. They also suggest the need
for a more refined notion of "economies of scope" that allows them to be partial rather than
complete. For example, our analysis in Table 7 shows virtually no competence-based
economies of scope between chemicals and computers, but a much higher probability of their
existence between chemicals and pharmaceutical products. A chemical firm would therefore
have partial economies of scope, ( in that it would probably still need to invest in certain
complementary fields of competence in order to enter pharmaceuticals) whereas in computers
it would be starting from scratch. In fact, the earlier concept of "technological relatedness"
(Rumelt, 1974) is analytically and operationally more useful, especially now that it can be
measured directly.
Explaining the Core - rather than the Boundaries - of the Firm. Our analysis suggests
that notions of firm-specific dynamic competencies provide a convincing empirical - and
competence-based - explanation of the core (not the boundaries) of the firm (Dosi et al.,
1992). In particular, our results in Tables 8 and 9 suggest that we are able to claim the
following: "Show us a firms' profile of technological competencies, and we shall probably be
able to predict the range of products it makes, and almost certainly be able to predict what it
does not make".
What Type of Variety? "Variety" is an often used term in evolutionary economics23. Our
analysis shows various possible definitions of the term, each of which should be carefully
distinguished in any general scheme of things:
Variety within firms in the technological competencies that they embody;
Variety between firms in their mix (or profile) of technological competencies, largely
defined by the products they develop and make.
a
Once given their profile of competencies and product mix, lack of variety in
managerial decisions about their directions of search, but considerable variety in
decisions about their rate of search.
5.3 For Future Research
Technology and Firm Performance. Continuing improvements in information technology
and firm-level data bases are opening considerable possibilities for econometric explorations
of the influence of technological activities on the performance of individual firms. Beyond
standard cross-sectional studies, attention should be paid to:
the effects of the level and field of education of management and the work force.
These could influence both the nature of decisions about the allocation of resources to
technology made under uncertainly, and the speed of learning (Bosworth and Wilson,
1992; Scherer and Huh, 1992; Prais, 1993; Pavitt and Patel, 1988)
the dynamic interactions between technological activities, product diversification and
performance.
23
See, for example, Metcalfe and Gibbons (1989).
Corporate Strategy and Technology Management. Qualitative studies (historical and casebased) will also be necessary to understand the dynamics of competence building and
exploitation in large firms. In particular:
the dynamic interactions in competence-accumulation amongst technologies,
components, sub-systems and products (see, for example, Abernathy and Clark, 1985;
Miyazaki, 1994);
the organisational implications of different technologies and types of innovation (see,
for example, Coombs and Richards, 1991; Tidd, 1993);
the problems of corporate technology strategy in the firm that is multi-product, multidivisional, and multi-technology (see, for example, Chandler, 1991, 1992);
the problems of defining corporate competencies ex ante rather than ex post.
A Red Herring. We should also avoid trying to answer unanswerable questions, such as why
particular firms choose to combine particular technologies in particular ways to make
particular products, from amongst the (almost) infinity of mathematical possibilities that do
exist. As in nature, firms evolve in a complex and path dependent world, where history
matters. If Darwin and DNA cannot model and predict the emergence and existence of the
elephant and the mouse, we should not be expected to do the equivalent in explaining why
industries are what they are and not something else.
REFERENCES
W. Abernathy and K. Clark (1985), "Mapping the Winds of Creative Destructive", Research
Policy, 14.
D. Archibugi (1988), "In Search of a Useful Measure of Technological Innovation",
Technological Forecasting and Social Change, 34.
A. Arora and A. Gambardella (1992), "New Trends in Technological Change: the Use of
General and Abstract Knowledge in Industrial Research", Rivista Internazionale di
Scienze Sociali, 3.
J. Barton (1993), "Adapting the Intellectual Property System to New Technologies", in
Wallerstein, Mogee and Schoen (eds.), Op. Cit.
B. Basberg (1987) "Patents and the Measurement of Technological Change: a Survey of the
Literature", Research Policy, 16.
M. Bell (1984), "Learning and the Accumulation of Technological Capacity in Developing
Countries", in M. Fransman and K. King (eds.), Technological Capability in the Third
World, Macmillan, London.
J. Bessant and W. Haywood (1988), "Islands, Archipelagos and Continents: Progress on the
Road to Computer Integrated Manufacturing", Research Policy, 17.
D. Bosworth and R. Wilson (1992), Technological Change: the Role of Scientists and
Engineers, Avebury, Aldershot.
Business Week (1993), "The Global Patent Race picks up Speed", August 9,49-54.
J. Cantwell (1989), Technological Innovation and Multinational Corporations, Blackwell,
Oxford.
J. Cantwell (1991), "The Theory of Technological Competence and its Application to
International Production", in D. McFetridge (ed.), Foreign Investment, Technology and
Economic Growth, University of Calgary Press, Calgary.
B. Carlsson and G. Eliasson (1991), The Nature and Importance of Economic Competence,
Working Paper, Swedish Board of Technical Development, Stockholm.
A. Chandler (1991), "The Functions of the HQ Unit in the Multibusiness firm", Strategic
Management Journal, 12.
A. Chandler (1992), "Corporate Strategy, Structure and Control Methods in the United States
during the 20th Century", Industrial and Corporate Change, 2.
W. Cohen and D. Levinthal (1989), "Innovation and Learning: the Two Faces of R & D",
Economic Journal, 99,569-596.
R. Coombs and A. Richards (1991), "Technologies, Products and Firm's Strategies",
Technology Analysis and Strategic Management, 3.
A. Cooper and D. Schendel (1976), "Strategic Responses to Technological Threats", Business
Horizons.
P. David (1975) Technical Choice, Innovation and Economic Growth, Cambridge University
Press, Cambridge.
G. Dosi, D. Teece and S. Winter (1992), "Towards a Theory of Corporate Coherence:
Preliminary Remarks", in G. Dosi, R. Giannetti and P. A. Toninelli (eds.), Technology
and Enterprise in a Historical Perspective, Clarendon Press, Oxford.
G. Dosi and D. Teece (1993), " Competencies and the Boundaries of the Firm", Center for
Research in Management, CCC Working Paper No. 93-1 1, University of California at
Berkeley.
European Industrial Research Management Association (EIRMA, 1986), Developing R & D
Strategies, Paris.
L. Franko (1989), "Global Corporate Competition: who's winning, who's losing, and the
R & D factor as one reason why". Strategic Management Journal, 10.
C. Freeman (1982), The Economics of Industrial Innovation, Pinter, London.
I. Galimberti (1993), Large Chemical Firms in Biotechnology: Case Studies of Learning in
Radically New Technology, D. Phil. Thesis, University of Sussex.
P. Geroski, S. Machin and J. van Reenen (1993), "The Profitability of Innovating Firms",
RAND Journal of Economics, 24.
0 . Granstrand and S. Sjolander (1990), "Managing Innovation in Multi-technology
Corporations", Research Policy, 19, 35-60.
0 . Granstrand, E. Bohlin , C. Oskarsson and N. Sjorberg (1992), "External Technology
Acquisition in Large Multi-technology Corporations", R & D Management, 22.
Z. Griliches (ed. 1984), Patents, R & D and Productivity, University of Chicago Press,
Chicago.
Z. Griliches (1990), "Patent Statistics as Economic Indicators", Journal of Economic
Literature, 28.
J. Hagedoorn and J. Schakenraad (1992), "Leading Companies in Networks of Strategic
Alliances in Information Technologies", Research Policy, 2 1.
B. Hall, Z. Griliches and A. Pakes (1986), "Patents and R & D: Is there a Lag", International
Economic Review, 27.
R. Henderson and I . Cockburn (1993), "Scale Scope and Spillovers: the Determinants of
Research Productivity in the Pharmaceutical Industry", NBER Working Paper No. 4466,
National Bureau of Economic and Social Research, Cambridge, MA.
E. von Hippel (1988), The Sources of Innovation, Oxford UP, New York.
M. Hobday (1993), "Technological Learning in Singapore: a Test Case of Leapfrogging",
Journal of Development Studies, 30,4, 83 1-858
T. Hughes and T. Pinch (1987) eds. The Social Construction of Technological Systems, MIT
Press Cambridge, Mass.
S. Jacobsson and C. Oskarsson (1993), "Educational Statistics as an Indicator of
Technological Activity", Research Policy, (forthcoming).
A. Jaffe (1986), "Technological Opportunity and Spillovers of R & D: Evidence from Firms',
Patents, Profits and Market Value", American Economic Review, 76, 984-1001.
A. Jaffe (1989), "Characterizing the 'Technological Position' of Firms, with Application to
Quantifying Technological Opportunity and Research Spillovers", Research Policy, 18,
87-97.
S. Kline (1990), "A Numerical Measure for the Complexity of Systems: the Concept and
some Implications", Report INN-5, Dept. of Mechanical Engineering, Stanford
University.
F. Kodama (1986a), "Technological Diversification of Japanese Industry", Science, 233 (18
July).
F. Kodama (1986b), "Japanese Innovation in Mechatronics Technology", Science and Public
Policy, 13.
C. Lindblom (1959), "The Science of Muddling Through", Public Administration Review, 19,
79-88.
F. Malerba (1992), "Learning by Firms and Incremental Technical Change", Economic
Journal, 102.
F. Malerba and L. Marengo (1993), "Competence, Innovative Activities and Economic
Performance in Italian High-Technology Firms", (mimeo), Bocconi University, Milan.
S. Metcalfe and M. Gibbons (1989), "Technology, Variety and Organisation: a Systematic
Perspective on the Competitive Process", Research on Technological Innovation,
Management and Policy, 4.
K. Miyazaki (1994), "Building Competencies in the Firm: Lessons from European and
Japanese Opt-electronics ", Macmillan (forthcoming).
D. Mowery (1988), ed. International Collaborative Ventures in US. Manufacturing,
Ballinger, Cambridge, Mass.
F. Narin, E. Noma and R. Perry (1987), "Patents as Indicators of Corporate Technological
Strength", Research Policy, 16.
F. Narin and D. Olivastro ((1988), "Technology Indicators based on Patents and Patent
Citations", in van Raan (ed), Op. Cit.
R. Nelson and S. Winter (1982), An Evolutionary Theory of Economic Change, Belknap,
Cambridge, Mass.
C. Oskarsson (1993), "Diversification and Growth in US, Japanese and European MultiTechnology Corporations", (mimeo), Dept. of Industrial Management and Economics,
Chalmers University of Technology, Gothenburg.
P. Pate1 (1993), "Localised Production of Technology for Global Markets", Cambridge
Journal of Economics, (forthcoming).
P. Pate1 and K. Pavitt (1987), "Is Western Europe losing the Technological Race?", Research
Policy, 16,59-85.
P. Pate1 and K. Pavitt (1991), "Large Firms in the Production of the World's Technology: an
Important Case of 'non-Globalisation' ", Journal of International Business Studies, 22,
1-21.
P. Pate1 and K. Pavitt (1992), "The Innovative Performance of the World's Largest Firms:
Some New Evidence", Economics of Innovation and New Technology, 2.91-102.
P. Patel and K. Pavitt (1994a), "Patterns of Technological Activity: their Measurement and
Interpretation", in P. Stoneman (ed.), Handbook of the Economics of Innovation and
Technical change, Blackwell, Oxford (forthcoming).
P. Pate1 and K. Pavitt (1994b), "The Continuing, Widespread (and Neglected) Importance of
Improvements in Mechanical Technologies", Research Policy (forthcoming).
K. Pavitt (1984), "Sectoral Patterns of Technical Change: Towards a Taxonomy and a
Theory", Research Policy, 13. 343-373.
K. Pavitt (1988), "Uses and Abuses of Patent Statistics" in van Raan, Op. Cit.
K. Pavitt and P. Patel (1988), "The International Distribution and Determinants of
Technological Activities", Oxford Review of Economic Policy, 4.
C. Perez and L. Soete (1988), "Catching up in Technology: Entry Barriers and Windows'of
Opportunity", in G. Dosi, C. Freeman, R. Nelson, G. Silverberg and L. Soete (eds.)
Technical Change and Economic Theory, Pinter, London.
M. Porter (1985), "Technology and Competitive Advantage", The Journal of Business
Strategy, 5.
M. Porter (1990), The Competitive Advantage of Nations, Macmillan, London.
C. Prahalad and G. Hamel (1990), "The Core Competence of the Corporation", Harvard
Business Review, (May-June), 79-9 1.
S. Prais, "Economic Performance and Education: the Nature of Britain's Deficiencies",
Discussion Paper No. 52, National Institute of Economic and Social Research, London.
J. Quinn (1980), Strategies for Change: Logical Incrementalism, Irwin, Homewood, Ill.
A. van Raan, ed., (1988), Handbook of Quantitative Studies of Science and Technology, North
Holland, Amsterdam.
V. Ramanujam and P. Varadarajan (1989), "Research on Corporate Diversification: a
Synthesis", Strategic Management Journal, 10, pp. 523-55 1.
N. Rosenberg (1976), "Technological Change in the Machine Tool Industry", Perspectives on
Technology, Cambridge UP, Cambridge.
R. Rosenbloom and M. Cusumano (1987), "Technological Pioneering and Competitive
Advantage: the Birth of the VCR Industry", California Management Review, 29.
R. Rumelt (1974), Strategy, Structure and Economic Performance, Graduate School of
Business Administration, Harvard University, Cambridge, Mass.
R. Rumelt, D. Schendel and D. Teece (1991), "Strategic Management and Economics", in
Strategic Management Journal, 12, pp. 5-29. Special Issue on Fundamental Research
Issues in Strategy and Economics.
P. Samuelson (1993) "A Case Study on Computer Programs", in Wallerstein, Mogee and
Schoen (eds.), Op. Cit.
F. Scherer (198 I), Industrial Market Structure and Economic Performance, Rand McNally.
F. Scherer and K. Huh (1992), "Top Management Education and R & D Investment",
Research Policy, 2 1.
M. Sharp (1991), 'LTechnologicalTrajectories and Corporate Strategies in the Diffusion of
Biotechnology", in E. Deiaco, E. Hornell and G. Vickery (eds.) Technology and
Investment, Pinter, London.
D. Teece, G. Pisano and A. Schuen (1992), "Dynamic Capabilities and Strategic
Management", Center for Research in Management, University of California at
Berkeley, CCC Working Paper.
D. Teece, R. Rumelt, G. Dosi and S. Winter (1993), "Understanding Corporate Coherence:
Theory and Evidence", Journal of Economic Behavior and Organization,
(forthcoming).
J. Tidd (1993), "Technological Innovation, Organizational Linkages and Strategic Degrees of
Freedom", Technology Analysis and Strategic Management, 5.
N. von Tunzelmann (1994), " 'Globalisation' of Technologies and the Electronics Industry",
mimeo, Science Policy Research Unit, Sussex University.
M. Tushman and P. Anderson (1987), "Technological Discontinuities and Organization
Environments", in A. Pettigrew (ed.), The Management of Strategic Change, Blackwell,
Oxford.
J. Utterback and F. Suarez (1993), "Innovation, Competition and Industrial Structure",
Research Policy, 22.
M. Wallerstein, M. Mogee and R. Schoen (eds. 1993), Global Dimensions of Intellectual
Property Rights in Science and Technology, National Academy Press, Washington, D.C.